Peisong Han is John G. Searle Assistant Professor in the Department of Biostatistics. Before joining the University of Michigan in 2018, Dr. Han was Assistant Professor in the Department of Statistics and Actuarial Science at the University of Waterloo, Canada, from 2013 to 2017. His primary research interests include (i) missing data and biased sampling problems in public health studies and survey sampling, (ii) data integration, especially when summary information is available for some studies and individual-level data are available for others, and (iii) longitudinal (correlated/clustered) data analysis. The main statistical methods he uses are empirical likelihood, estimating equations, and James-Stein shrinkage. Interdisciplinary collaborations include bipolar disorder research, food insecurity and nutrition incentive program research, and kidney disease quality measure development, maintenance and support.
PhD, Biostatistics, University of Michigan, 2013
M.S., Statistics, Michigan State University, 2008
B.S., Mathematics, University of Science and Technology of China, 2006
Research Interests & Projects
- Missing data and biased sampling problems. I have been developing a systematic estimation procedure that can simultaneously accommodate multiple working models for both the missingness probability and the data distribution. The resulting estimators are “unbiased” if any one working model is correctly specified and are “good” if no working model is correct. This research is especially relevant in the presence of high dimensional covariates when working model specification is challenging. An R package “MultiRobust” is available on CRAN. The ideas can be used to correct for the selection bias for certain biased sampling problems.
- Data integration. The overarching goal is to incorporate external data/information into fitting models for an “internal” study for potential estimation efficiency gains. I have been working on different projects that cover a variety of different concrete settings. For example, a couple of projects involve using information from existing risk or survival probability calculators for certain diseases of interest where the calculators are a “black box”. A few other projects deal with population heterogeneity when integrating data from different sources, an important yet challenging setting.
- Bipolar disorder research. This is in collaboration with the Heinz C. Prechter Bipolar Research Program at Michigan Medicine. The goals are to discover the fundamental biological changes that cause bipolar disorder and develop new interventions to treat and prevent the illness. One project focuses on the temporal stability of personality traits among individuals with bipolar disorder and the association between bipolar disorder diagnosis and the longitudinal trajectories of personality traits and moods. Another project investigates the short and long term impacts of COVID-19 pandemic on individuals with BD.
- Kidney disease quality measure. This is in collaboration with the U-M Kidney Epidemiology and Cost Center (KECC). I am involved in several projects on developing, maintaining and updating the technical specifications for current and future quality measures for chronic kidney disease and end-stage renal disease (ESRD).
- Food insecurity and nutrition intake. This is in collaboration with the Gus Schumacher Nutrition Incentive Program funded by USDA through the Gretchen Swanson Center for Nutrition in Nebraska. Projects focus on improving dietary health of low-income individuals, reducing individual and household food insecurity, and how nutrition incentive programs can help achieve these goals.
- Han, P., Kong, L., Zhao, J., and Zhou, X. (2019) A General Framework for Quantile Estimation with Incomplete Data. Journal of the Royal Statistical Society - Series B. 81, 305-333.
- Han, P., and Lawless, J. F. (2019) Empirical Likelihood Estimation Using Auxiliary Summary Information with Different Covariate Distributions. Statistica Sinica. 29, 1321-1342.
- Han, P. (2018). A Further Study of Propensity Score Calibration in Missing Data Analysis. Statistica Sinica, 28, 1307-1332.
- Han, P. (2018). Calibration and Multiple Robustness When Data Are Missing Not At Random. Statistica Sinica, 28, 1725-1740.
- Han, P. (2016). Intrinsic Efficiency and Multiple Robustness in Longitudinal Studies with Dropout. Biometrika. 103, 683-700.
- Han, P. (2016). Combining Inverse Probability Weighting and Multiple Imputation to Improve Robustness of Estimation. Scandinavian Journal of Statistics, 43, 246-260.
- Han, P., Song, P. and Wang, L. (2015). Achieving Semiparametric Efficiency Bound in Longitudinal Data Analysis with Dropouts. Journal of Multivariate Analysis, 135, 59-70.
- Han, P. (2014). Multiply Robust Estimation in Regression Analysis with Missing Data. Journal of the American Statistical Association, 109, 1159-1173.
- Han, P. (2014). A Further Study of the Multiply Robust Estimator in Missing Data Analysis. Journal of Statistical Planning and Inference, 148, 101-110.
- Han, P. and Wang, L. (2013). Estimation with Missing Data: Beyond Double Robustness. Biometrika, 100 (2), 417-430.